- Email: [email protected]
Ethnicity and Depressive Symptoms in Latino and White Patients with Coronary Artery Disease
o
r
i
g
i
n
a
l
c
o
m
m
u
n
i
c
a
t
i
o
n
Psychosocial Mediators of the Relationship between Race/Ethnicity and Depressive Symptoms in Latino and White Patients with Coronary Artery Disease C. Boutin-Foster, MD; G. Ogedegbe, MD; J. Peterson, EdD; W.M. Briggs, PhD; J.P. Allegrante, PhD; and M.E. Charlson, MD
Financial support: This study was supported by 1R01 HL62161 from the National Heart, Lung and Blood Institute. Background: The high prevalence of depressive symptoms in patients with coronary artery disease has been well documented. However, little is known about the prevalence and correlates of depressive symptoms in Latino patients with coronary artery disease. Purpose: Among Latino and white patients who had percutaneous transluminal coronary angioplasty (PTCA), this study examined whether differences in the prevalence of depressive symptoms exist and the degree to which psychosocial factors (years of education, employment status, stressful life events, emotional social support) explained any differences. Methods: Using a cross-sectional design, closed-format questionnaires were used to obtain clinical and psychosocial history. The definition of high depressive symptoms was based on a score of ≥16 on the Center for Epidemiologic Studies Depression Scale (CES-D). Results: Compared to whites (n=492), Latinos (n=59) were younger, and a greater proportion were female, but fewer completed high school and fewer were employed (P<0.05). More Latinos reported experiencing ≥2 recent stressful life events, but fewer reported having emotional social support (P<0.05). There was a significant association between race/ethnicity and depressive symptoms (OR=2.3, 95% CI: 1.3–4.5). In multivariate analyses, the significance of this association diminished when psychosocial variables were added to the model. Conclusions: In this study, education, employment, stressful life events and emotional social support partially explained the observed racial/ethnic differences in depressive symptoms. Key words: race/ethnicity n coronary artery disease n depression n Latinos
JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
© 2008. From the Center for Complementary and Integrative Medicine, Weill Medical College, Cornell University, New York, NY (Boutin-Foster, Peterson, Briggs, Charlson); and Department of Medicine, College of Physicians and Surgeons (Ogedegbe), Department of Health and Behavior Studies, Teachers College, and Department of Sociomedical Sciences, Mailman School of Public Health, Columbia University, New York, NY (Allegrante). Send correspondence and reprint requests for J Natl Med Assoc. 2008;100:849–855 to: Dr. Carla Boutin-Foster, , Weill Medical College, Cornell University, 525 E. 68th St., Room F1421, Box 46, New York, NY 10021; e-mail: [email protected]
Introduction
S
everal epidemiological studies have examined the mental health status of Latino residents living in the United States. These studies include the National Institute of Mental Health (NIMH) Epidemiological Catchment Area Program,1 the Hispanic Health and Nutrition Examination Survey2 and the Hispanic Established Populations for the Epidemiologic Studies of the Elderly.3 Most of these studies were conducted among Latinos of Mexican, Cuban or Puerto-Rican descent and conducted in community-based settings. The most common measures of depressive symptoms employed in these studies included the Center for Epidemiologic Studies Depression Scale (CES-D), the Composite International Diagnostic Interview, the NIMH Diagnostic Interview Schedule and the Beck Depression Inventory.4 Irrespective of the country of origin or the measurement used, the consensus of these studies is that Latinos experience a high prevalence of depressive symptoms and that the prevalence of depressive symptoms observed among some Latinos is higher than that observed among non-Hispanic or Latino whites.4 The prevalence of depressive symptoms in these studies varied from 3–28%.5 For example, a survey of Cuban Americans living in Florida reported a lifetime prevalence of high depressive symptoms of 3%.5 Another study conducted in a cohort of community dwellers in Los Angeles County found a 27% prevalence of depresVOL. 100, NO. 7, JULY 2008 849
Race/Ethnicity and Depressive Symptoms in CAD Patients
sive symptoms among Latinos compared to a 16% prevalence in whites.6 The prevalence among Puerto Ricans living in New York City was found to be 28%.2 Several studies have also reported more adverse outcomes in patients with depressive symptoms compared to those without symptoms. Variables that have been shown to correlate the most with depressive symptoms include younger age, female gender, fewer years of education, unemployment and lower income.4 While there is ample evidence documenting the high prevalence of depressive symptoms in Latinos, there is little research on patients who have both depressive symptoms and comorbid medical conditions. One condition that deserves further investigating is coronary artery disease, the leading cause of death among Latino adults. The proportion of premature deaths attributable to coronary artery disease is estimated at 23% in Latino adults compared to 14% in white adults.7 Coronary artery disease in the Latino population is of particular concern given their high burden of cardiovascular risk factors coupled with limited socioeconomic resources and limited access to healthcare.8 Another risk factor that is understudied but that may have an equally deleterious impact on the health outcomes of Latinos with coronary artery disease is depressive symptoms. In the general population, approximately 20% of patients with coronary artery disease (CAD) are diagnosed with depressive symptoms, with 14% of patients developing new onset depressive symptoms within a year of a cardiac event.9,10 Moreover, patients with depressive symptoms are more likely to develop recurrent infarctions and more likely to undergo repeat coronary procedures than those without such symptoms.11-16 In the Hispanic Established Population for the Epidemiologic Study of the Elderly (EPESE), Latino patients who had comorbid cardiovascular disease and depressive symptoms had a mortality rate that was four times higher than patients with cardiovascular disease but who did not have depressive symptoms.17 Given the high prevalence of depressive symptoms among Latinos and the known adverse outcomes associated with such symptoms, especially in patients with coronary artery disease, depressive symptoms in Latino patients with CAD require additional research. Therefore, among Latino and white patients who recently had percutaneous transluminal coronary angioplasty (PTCA), the objectives of this study were two-fold: 1) to determine whether differences in the prevalence of depressive symptoms exist; and 2) to determine the degree to which years of education, employment status, experience of stressful life events and perceived emotional social support explained any observed differences. The hypotheses for this study are: 1) the prevalence of depressive symptoms would be higher among Latino patients compared to white patients and 2) the differences in depressive symptoms noted between Latino and white patients would be explained 850 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
by differences in education, employment status, recent stressful life events and perceived emotional social support. The rationale for these hypotheses is that there is considerable evidence to suggest that racial/ethnic categories may shape social experiences such as education, employment, social interactions and exposure to stressful events. These factors may in turn contribute the higher prevalence of depressive symptoms that has been reported among racial and ethnic minorities.18,19
Methods
This study was nested within a larger randomized controlled trial designed to improve behavior modification among patients who recently had coronary artery angioplasty. The methodologic details of the larger study have been described elsewhere.20
Patient Sample Patients were recruited from a tertiary care hospital telemetry unit and had to have angiographic evidence of coronary artery disease warranting either angioplasty and/ or stent insertion. Eligible patients who provided consent underwent baseline interviews that included an assessment of demographic characteristics (age, gender and ethnicity/race, years of education and employment status), clinical characteristics (severity of cardiac disease, medical comorbidity and quality of life) and psychosocial characteristics (depressive symptoms, perceived emotional social support and recent stressful life events).
Measures Severity of cardiac disease was measured with the Canadian Cardiovascular Society Classification system, which rates the severity of anginal symptoms from the mildest anginal symptoms (class I) to most severe symptoms (class IV).21 In addition, perceived quality of life was measured with the Medical Outcomes Study-SF36, which measures patients’ perceptions of their overall health and quality of life.22 Depressive symptoms were measured with the Center for Epidemiologic Studies Depression Scale (CESD), a 20-item measure of affective and somatic depressive symptoms with scores ranging from 0–60. A score of ≥16 has been previously validated as being indicative of having high depressive symptoms.23 Social support items explored the amount of emotional support the patient perceived such as having a friend with whom to share important and personal matters and the degree of social contact the patient had with social networks.24 Patients rated the amount of time that they had someone to talk to about personal matters from none of the time to a lot. This single-item method of assessing emotional social support has been used in prior studies.25,26 Stressful life events in the year prior to angioplasty were documented through the use of questions about family stressors, work stressors, personal stressors VOL. 100, NO. 7, JULY 2008
Race/Ethnicity and Depressive Symptoms in CAD Patients
and financial stressors. For example, participants were asked whether they had experienced the death of a family member or friend, a divorce or separation, financial troubles, troubles with children or grandchildren, a mugging, robbery, or similar events during the previous year.27 Patients were given a checklist from which they were able to select stressful life events experienced. The checklist approach allowed patients to select >1 category. Similar methodology has been used to assess stressful life events in prior studies of Latino patients.26
Data Analyses Latino categorization was based on patients’ selfidentification as either Latino or Hispanic and was contrasted with patients who identified themselves as white and non-Hispanic or non-Latino. Demographic and clinical characteristics of Latinos and whites were compared and contrasted using Chi-squared test for categorical variables and Student’s t test for continuous variables. Patients who had a score of ≥16 on the CES-D were categorized as having high depressive symptoms compared to those with a score of <16, who were categorized as having low depressive symptoms. In order to address the hypothesis that Latino patients have a higher prevalence of depressive symptoms, the proportion of depressive symptoms by race was determined. Chi square was used as a test of difference in proportions. The analytic framework for testing mediators proposed by Baron and Kenny was used to address the hypothesis that observed differences in depressive symptoms would be explained by psychosocial factors.28 Psychosocial variables that were known to be associated with both race/ethnicity and depressive symptoms were identified and tested to determine the extent to which these variables mediated the association between race/ ethnicity and depressive symptoms. According to Baron and Kenny, four conditions must be met for a variable to be considered a mediator: 1) the independent variable (race/ethnicity) must be associated with the dependent variable (high depressive symptoms); 2) the independent variable must be associated with the proposed mediators (education, employment, perceived emotional social support and recent stressful life events); 3) the proposed mediators must also be associated with the dependent variable; and 4) A test for mediation holds when the magnitude of the association between the independent and dependent variable diminishes when the proposed mediators are added to the model. The first three criteria were established with bivariate analyses. Intercorrelations between race/ethnicity and depressive symptoms, race/ethnicity and potential mediators, and potential mediators and depressive symptoms were tested with Pearson correlation. The fourth criterion was established with multivariate logistic regression models that adjusted for age, gender, severity of illness and quality of life. In each model, depressive symptoms indicated JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
by a CES-D score of ≥16 or not was the dependent variable, and race/ethnicity was the independent variable. All tests were performed using SAS version 8.29 Because many of the variables in these models were interrelated, the variance inflation factors (VIFs) for each variable was determined to test for potential of multicollinearity affecting the results. VIF measures the degree to which collinearity among the independent variables in a regression model biases associations. This was done by entering all possible independent variables into one model and calculating the VIF. A VIF of >10 indicates that multicollinearity is a problem.30 None of VIFs were >10 in these analyses.
Results
A total of 2,022 patients were identified and screened for potential eligibility through a systematic, daily review of the cardiac catheterization schedule during an 18-month period. Patients had to be admitted for PTCA or stenting. Of these patients, 1,429 were eligible, of which approximately 46% were enrolled. Patients who were eligible but were not enrolled were those who either refused or those who had a family member or physician that did not allow them to participate. There were also patients who were not enrolled because they were admitted and discharged before they could be recruited. Of the patients who were eligible but not enrolled, 13% were Latino, their mean age was 62 ± 10, and 41% were female. The remainder of the paper will focus on the 660 patients who were enrolled. Of the 660 patients who were enrolled in the parent study, 551 (59 Latino patients and 492 white patients) were included in these analyses. Table 1 shows a comparison of demographic and clinical characteristics between Latino and white patients. Latino patients were younger than white patients and had a mean age of 56 ± 10 compared to 63 ± 11 (P=0.001). Of the Latino patients, 42% were female compared to 25% of white patients, (P=0.007). Compared to white patients, fewer Latino patients were employed (P=0.001) or completed high school (P=0.001). More Latinos than whites had a history of diabetes (P= 0.003) and rated their health as fair or poor (P=0.001) on the SF-36. The mean score on the CES-D among whites was 12.6 ±11, compared to 19.2 ±15 among Latino patients (P=0.002). As shown in Table 2, 60% of Latino patients had CES-D scores of ≥16, compared to 29% of whites. Latinos were >3 times as likely to have high depressive symptoms (OR=3.5, 95% CI: 2.0–6.1). Latino patients were less likely to report that they had someone to talk to about important matters in their lives some or most of the time, compared to whites (OR=0.33, 95% CI: 15– 70). The majority of participants reported experiencing ≥1 stressful life event. Therefore, patients who reported ≥2 life events were compared with those who reported fewer events. Latinos were >3 times as likely to report VOL. 100, NO. 7, JULY 2008 851
Race/Ethnicity and Depressive Symptoms in CAD Patients
that they experienced ≥2 recent stressful life events, (OR=3.6, 95% CI: 1.9–6.9). The first three criteria of Baron and Kenny’s model were established via bivariate analyses. Pearson correlation coefficients demonstrated that the associations among race/ethnicity and depressive symptoms, race/ ethnicity and the potential mediators, and the potential mediators and depressive symptoms were all significantly correlated at the p<0.05 significance level. The final criterion was established in the multivariate models shown in Table 3, which demonstrates the impact of separately adding each of the proposed mediators on the association between race/ethnicity and depressive symptoms. As shown in Table 3, race/ethnicity was associated with depressive symptoms (OR=2.3, CI: 1.3–4.5). When the variable for recent stressful life events was added to the model, the odds ratio for the association between race/ethnicity and depressive symptoms decreased from 2.3 to 2.2 (95% CI: 1.2–4.2). When social support was added to the model as a mediator, the odds ratio for the association between race/ethnicity and depressive symptoms decreased to 2.1 (95% CI: 1.1–9.0). When education was added to the model as a mediator, the odds ratio for the association between race/ethnicity and depressive symptoms decreased to 2.0 (95% CI: 1.1–4.0). However, when employment was added to the model, the association between race/ethnicity and depressive symptoms was no longer significant—the odds ratio became 1.79 (95% CI: 0.75–4.2). All models included age, gender, severity of illness and quality of life.
Discussion
This study applied Baron and Kenny’s framework to
identify potential mediators in the association between race/ethnicity and depressive symptoms. Of the Latino patients in this study, 60% had high depressive symptoms, compared to 29% of whites, which established the criteria in Baron and Kenny’s model that the independent variable must be associated with the dependent variable. Bivariate associations between race/ethnicity and the potential mediators and between the potential mediators and depressive symptoms were also established. The final criterion for mediation was confirmed in logistic regression models that showed that the association between racial/ethnicity and depressive symptoms was partially explained by fewer years of education, being unemployed, lacking emotional social support and experiencing ≥2 recent stressful life events. Of these variables, unemployment had the greatest impact on the association between race/ethnicity and depressive symptoms. This is relevant because employment status may influence access to care and the ability to manage one’s health. This is especially important since Latinos in this study were younger, more likely to be female and unmarried. In this context, employment status would have an important impact on the patients’ ability to care for themselves and perhaps their family. The fact that there were more female participants among Latinos compared to whites is worth further mention. It has been documented that there is a greater prevalence of depressive symptoms among women than men in both the general population and in Latino populations.31,32 It is also known that among patients with coronary artery disease, women are twice as likely to have depressive symptoms.33 Although gender did not affect the association between depressive symptoms and race/ ethnicity, depressive symptoms in Latina women are of
Table 1. Clinical characteristics of Latino patients compared to white patients
Demographic Characteristics Age (years) + SD Female Married Employed High-school graduate Clinical History Diabetes Self-rated health as fair or poor Hypertension history Unstable angina Previous infarction Previous heart failure history Severity of illness (CCS) class IV*
Latino (n=59)
White (n=601)
P
56 ± 10 42% 56% 44% 58%
63 ± 11 25% 63% 56% 88%
0.001 0.007 0.001 0.001 0.001
43% 64% 58% 66% 27% 15% 57%
24% 29% 54% 50% 24% 10% 47%
0.003 0.001 NS* NS NS NS NS
* Canadian Cardiovascular Society Classification; ** NS: Nonsignificant, indicates that p>00.05
852 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
VOL. 100, NO. 7, JULY 2008
Race/Ethnicity and Depressive Symptoms in CAD Patients
particular concern. According to traditional sex roles that define many Latino households, it is common for Latina women to have to tend for the household needs of the family such as caring for the children and their partner.34 Latina females in this study may have had to maintain their roles in the family while managing both coronary artery disease and depressive symptoms. Future studies focusing on depressive symptoms among Latina women may help to further explore this assumption. These findings must also be viewed in light of what is known about the socioeconomic status of Latinos living in the United States. According to the Department of Health and Human Services 2000 Census report, some subgroups within the Latino population may have fewer socioeconomic resources in terms of years of education, income and employment that are available to them.35 Education and employment are economic resources that are necessary to cope effectively with day-to-day stressors. Thus, patients who lack or are bereft of such resources may be at greater risk for depressive symptoms.36 The finding that Latino patients reported lower perceived social support is surprising, given that social support—in particular, family support—has consistently been shown to be a protective factor within Latino families.26 The cultural value of familismo has been used to describe the strong attachments to extended and nuclear families.37,38 However, the role of social support in patients with comorbid illnesses is unknown. It is possible that the larger family networks may not be able to adequately address the needs of family members who have comorbid medical conditions, especially if socioeconomic resources are limited. While there may be other limitations in this study, we
will focus on four limitations that were felt to be most germane to interpreting and extrapolating our results. These limitations occur in terms of the number of Latinos recruited, the selection of study participants, the measurement of social support and stressful life events and the method used to test for mediation. First, there were too few Latinos to allow for within Latino comparisons of those patients who had high depressive symptoms and those that did not. Second, common to studies in clinically predefined populations, the selection of study participant limits the ability to generalize the findings to other Latinos who do not have coronary artery disease. The requirement that patients spoke English also limits the generalizability of these findings and the number of Latino participants in the study. Future studies can address these limitations and build upon our findings by recruiting a larger number of Latinos patients with other comorbid conditions, who have diverse language fluency, and by also capturing information on their country of origin and level of acculturation. This would allow for a more in-depth description of the tremendous heterogeneity that exists within the Latino population. Third, the use of a single item to measure emotional social support is another limitation of the study. Compared to multi-item measures, a single item precludes the ability to evaluate other aspects of social support such as informational and tangible support. However, of all of the categories of social support, emotional social support has consistently been shown to predict health outcomes. Previous studies have used a similar item to measure emotional social support and found it be predictive of mortality after an infarction, fatal and nonfatal events in the setting of heart failure and intention to participate
Table 2. Psychosocial characteristics by race/ethnicity
Psychosocial Characteristics Scored ≥16 on the CES-D Had someone to talk to some or most of the time Number of recent life events
Latino (n=59) 60% 81% 28%
White (n=601) 29% 92% 9%
Odds Ratio for Latino (95% CI) 3.5 (2.0,6.1) 0.33 (0.15, 0.70) 3.6 (1.9,6.9)
Table 3. Multivariate models demonstrating the impact of adding mediators on the association between race/ethnicity on depressive symptoms
Model Race/ethnicity Stressful life event Emotional social support Education Employment
Odds Ratio
95% CI
2.3 2.2 2.1 2.0 1.8
1.3-4.5 1.2-4.2 1.1-9.0 1.1-4.0 0.75-4.2
* All models adjusted for age, gender, severity of illness.
JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
VOL. 100, NO. 7, JULY 2008 853
Race/Ethnicity and Depressive Symptoms in CAD Patients
in healthy behaviors.25,39,40 The checklist approach to measuring life events has been commonly used in prior studies, but it does not allow for a more detailed evaluation of the specific categories of events. In this study, some respondents experienced >1 event, and this approach allowed them to select >1 stressful event. However, because the duration of the event was not probed, the impact that duration of life events may have had on depressive symptoms could not be determined. This is an important consideration for future studies since the presence of chronic stress may represent unresolved issues that continually cause distress. Finally, a series of logistic regression models was used to test for mediation using Baron and Kenny’s approach as a framework. This is a valid method; however, it limited the ability to identify direct and indirect pathways leading to symptoms of depressive symptoms. Despite these limitations, this study contributes to an understanding of ways in which psychosocial variables may influence mental health in Latino patients with coronary artery disease. Examining these variables in the context of coronary artery disease is important in light of three contemporary trends that have been observed. These are: the increasing number of Latinos in the United States; the increasing prevalence of depressive symptoms in the general population; and the increasing prevalence in cardiovascular risk factors, including obesity, physical inactivity and type-2 diabetes.41-44 These trends make it highly probable that an increase in both the prevalence of depressive symptoms and coronary artery disease will be observed in the Latino population. This probability is further complicated by the lower use of mental health services among Latinos.45,46 Therefore, Public health initiatives to screen for depression and to provide referrals are important to addressing mental health in Latinos, in particular, those with coronary artery disease. In conclusion, our findings do not suggest that Latinos have an inherent negative disposition but, rather, illustrate the importance of understanding the factors that may influence the presence of depressive symptoms. Race/ethnicity may be a proxy measure for a set of interrelated social experiences such as access to education, employment, social interactions and exposure to stressful events. Reducing racial/ethnic disparities in mental health may require public health initiatives that address these psychosocial factors.
References
1. Golding JM, Lipton RI. Depressed mood and major depressive disorder in two ethnic groups. J Psychiatr Res. 1990;24(1):65-82. 2. Potter LB, Rogler LH, Moscicki EK. Depression among Puerto Ricans in New York City: the Hispanic Health and Nutrition Examination Survey. Soc Psychiatry Psychiatr Epidemiol. 1995;30(4):185-193. 3. Black SA, Markides KS, Miller TQ. Correlates of depressive symptomatology among older community-dwelling Mexican Americans: the Hispanic EPESE. J Gerontol B Psychol Sci Soc Sci. 1998;53(4):S198-208. 4. Saez-Santiago E, Bernal G. Depression in ethnic minorities: Latinos and
854 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
Latinas, African Americans, Asian Americans, and Native Americans. Handbook of racial & ethnic minority psychology. Vol 4. Thousand Oaks, CA: Sage; 2003:401-428. 5. Narrow WE, Rae DS, Moscicki EK, et al. Depression among Cuban Americans. The Hispanic Health and Nutrition Examination Survey. Soc Psychiatry Psychiatr Epidemiol. 1990;25(5):260-268. 6. Frerichs RR, Aneshensel CS, Clark VA. Prevalence of depression in Los Angeles County. Am J Epidemiol. 1981;113(6):691-699. 7. Thom T, Haase N, Rosamond W, et al. Heart disease and stroke statistics—2006 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2006;113(6): e85-e151. 8. Sundquist J, Winkleby MA, Pudaric S. Cardiovascular disease risk factors among older black, Mexican-American, and white women and men: an analysis of NHANES III, 1988–1994. Third National Health and Nutrition Examination Survey. J Am Geriatr Soc. 2001;49(2):109-116. 9. Mallik S, Krumholz HM, Lin ZQ, et al. Patients with depressive symptoms have lower health status benefits after coronary artery bypass surgery. Circulation. 2005;111(3):271-277. 10. Lauzon C, Beck CA, Huynh T, et al. Depression and prognosis following hospital admission because of acute myocardial infarction. CMAJ. 2003;168(5):547-552. 11. Borowicz L, Royall R, Grega M, et al. Depression and cardiac morbidity 5 years after coronary artery bypass surgery. Psychosomatics. 2002;43(6):464471. 12. Burker EJ, Blumenthal JA, Feldman M, et al. Depression in male and female patients undergoing cardiac surgery. Br J Clin Psychol. 1995;34(Pt 1):119-128. 13. Carney RM, Freedland KE, Veith RC, et al. Major depression, heart rate, and plasma norepinephrine in patients with coronary heart disease. Biol Psychiatry. 1999;45(4):458-463. 14. Carney RM, Freedland KE, Eisen SA, et al. Major depression and medication adherence in elderly patients with coronary artery disease. Health Psychol. 1995;14(1):88-90. 15. Carney RM, Rich MW, Freedland KE, et al. Major depressive disorder predicts cardiac events in patients with coronary artery disease. Psychosom Med. 1988;50(6):627-633. 16. Musselman DL, Evans DL, Nemeroff CB. The relationship of depression to cardiovascular disease: epidemiology, biology, and treatment. Arch Gen Psychiatry. 1998;55(7):580-592. 17. Black SA, Markides KS. Depressive symptoms and mortality in older Mexican Americans. Ann Epidemiol. 1999;9(1):45-52. 18. Turner RJ, Lloyd DA. The stress process and the social distribution of depression. J Health Soc Behav. 1999;40(4):374-404. 19. Schulz AJ, Israel BA, Zenk SN, et al. Psychosocial stress and social support as mediators of relationships between income, length of residence and depressive symptoms among African American women on Detroit’s eastside. Soc Sci Med. 2006;62(2):510-522. 20. Charlson ME, Allegrante JP, McKinley PS, et al. Improving health behaviors and outcomes after angioplasty: using economic theory to inform intervention. Health Educ Res. 2002;17(5):606-618. 21. Campeau L. The Canadian Cardiovascular Society grading of angina pectoris revisited 30 years later. Can J Cardiol. 2002;18(4):371-379. 22. Ware JE, Kosinski M, and Keller SD. SF-36 Physical and Mental Health Summary Scales: A User’s Manual. Boston: The Health Institute, New England Medical Center; 1994. 23. Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385-401. 24. Gorkin L, Schron EB, Brooks MM, et al. Psychosocial predictors of mortality in the Cardiac Arrhythmia Suppression Trial-1 (CAST-1). Am J Cardiol. 1993;71(4):263-267. 25. Berkman LF, Leo-Summers L, Horwitz RI. Emotional support and survival after myocardial infarction. A prospective, population-based study of the elderly. Ann Intern Med. 1992;117(12):1003-1009. 26. Chiriboga DA, Black SA, Aranda M, et al. Stress and depressive symptoms among Mexican American elders. J Gerontol B Psychol Sci Soc Sci. 2002;57(6):P559-568. 27. Ahern DK, Gorkin L, Anderson JL, et al. Biobehavioral variables and mor-
VOL. 100, NO. 7, JULY 2008
Race/Ethnicity and Depressive Symptoms in CAD Patients tality or cardiac arrest in the Cardiac Arrhythmia Pilot Study (CAPS). Am J Cardiol. 1990;66(1):59-62. 28. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality & Social Psychology. 1986;51(6):1173-1182. 29. SAS Institute Inc., SAS OnlineDoc®, Version 8, Cary, NC: SAS Institute Inc., 2000. 30. Tucker JL. The moderators of patient satisfaction. J Manag Med. 2002;16:48-66. 31. Salgado de Snyder VN, Cervantes RC, Padilla AM. Gender and ethnic differences in psychosocial stress and generalized distress among Hispanics. Sex Roles. 1990;22(7-8):441-453. 32. Koss-Chioino JD. Depression among Puerto Rican Women: Culture, Etiology and Diagnosis. Hispanic Journal of Behavioral Sciences. 1999;21(3):330-350. 33. Frasure-Smith N, Lesperance F, Juneau M, et al. Gender, depression, and one-year prognosis after myocardial infarction. Psychosom Med. 1999;61(1):26-37. 34. Miranda AO, Bilot JM, Peluso PR, et al. Latino Families: the Relevance of the Connection Among Acculturation, Family Dynamics, and Health for Family Counseling Research and Practice. The Family J. 2006;14(3): 268273. 35. Ramirez R. We the People: Hispanics in the United States. www.census. gov/prod/2004pubs/censr. Accessed 05/06/08. 36. Plant EA, Sachs-Ericsson N. Racial and ethnic differences in depression: the roles of social support and meeting basic needs. J Consult Clin Psychol. 2004;72(1):41-52. 37. La Roche MJ. The Association of Social Relations and Depression Levels among Dominicans in the United States. Hispanic J Behav Sci. 1999;21(4):420-430.
c
a
r
e
e
r
o
p
p
o
r
JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
t
38. Losada A, Robinson Shurgot G, Knight BG, et al. Cross-cultural study comparing the association of familism with burden and depressive symptoms in two samples of Hispanic dementia caregivers. Aging Ment Health. 2006;10(1):69-76. 39. Rhodes RE, Jones LW, Courneya KS. Extending the theory of planned behavior in the exercise domain: a comparison of social support and subjective norm. Res Q Exerc Sport. 2002;73(2):193-199. 40. Krumholz HM, Butler J, Miller J, et al. Prognostic importance of emotional support for elderly patients hospitalized with heart failure. Circulation. 1998;97(10):958-964. 41. Larkin GL, Claassen CA, Emond JA, et al. Trends in U.S. emergency department visits for mental health conditions, 1992 to 2001. Psychiatr Serv. 2005;56(6):671-677. 42. Klerman GL, Weissman MM. Increasing rates of depression. JAMA. 1989;261(15):2229-2235. 43. Arnett DK, McGovern PG, Jacobs DR, et al. Fifteen-year trends in cardiovascular risk factors (1980-1982 through 1995-1997): the Minnesota Heart Survey. Am J Epidemiol. 2002;156(10):929-935. 44. Lorenzo C, Okoloise M, Williams K, et al. The metabolic syndrome as predictor of type 2 diabetes: the San Antonio heart study. Diabetes Care. 2003;26(11):3153-3159. 45. Alegria M, Canino G, Rios R, et al. Inequalities in use of specialty mental health services among Latinos, African Americans, and non-Latino whites. Psychiatr Serv. 2002;53(12):1547-1555. 46. Miranda J, Cooper LA. Disparities in care for depression among primary care patients. J Gen Intern Med. 2004;19(2):120-126. n
u
n
i
t
i
e
s
VOL. 100, NO. 7, JULY 2008 855
r
i
g
i
n
a
l
c
o
m
m
u
n
i
c
a
t
i
o
n
Psychosocial Mediators of the Relationship between Race/Ethnicity and Depressive Symptoms in Latino and White Patients with Coronary Artery Disease C. Boutin-Foster, MD; G. Ogedegbe, MD; J. Peterson, EdD; W.M. Briggs, PhD; J.P. Allegrante, PhD; and M.E. Charlson, MD
Financial support: This study was supported by 1R01 HL62161 from the National Heart, Lung and Blood Institute. Background: The high prevalence of depressive symptoms in patients with coronary artery disease has been well documented. However, little is known about the prevalence and correlates of depressive symptoms in Latino patients with coronary artery disease. Purpose: Among Latino and white patients who had percutaneous transluminal coronary angioplasty (PTCA), this study examined whether differences in the prevalence of depressive symptoms exist and the degree to which psychosocial factors (years of education, employment status, stressful life events, emotional social support) explained any differences. Methods: Using a cross-sectional design, closed-format questionnaires were used to obtain clinical and psychosocial history. The definition of high depressive symptoms was based on a score of ≥16 on the Center for Epidemiologic Studies Depression Scale (CES-D). Results: Compared to whites (n=492), Latinos (n=59) were younger, and a greater proportion were female, but fewer completed high school and fewer were employed (P<0.05). More Latinos reported experiencing ≥2 recent stressful life events, but fewer reported having emotional social support (P<0.05). There was a significant association between race/ethnicity and depressive symptoms (OR=2.3, 95% CI: 1.3–4.5). In multivariate analyses, the significance of this association diminished when psychosocial variables were added to the model. Conclusions: In this study, education, employment, stressful life events and emotional social support partially explained the observed racial/ethnic differences in depressive symptoms. Key words: race/ethnicity n coronary artery disease n depression n Latinos
JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
© 2008. From the Center for Complementary and Integrative Medicine, Weill Medical College, Cornell University, New York, NY (Boutin-Foster, Peterson, Briggs, Charlson); and Department of Medicine, College of Physicians and Surgeons (Ogedegbe), Department of Health and Behavior Studies, Teachers College, and Department of Sociomedical Sciences, Mailman School of Public Health, Columbia University, New York, NY (Allegrante). Send correspondence and reprint requests for J Natl Med Assoc. 2008;100:849–855 to: Dr. Carla Boutin-Foster, , Weill Medical College, Cornell University, 525 E. 68th St., Room F1421, Box 46, New York, NY 10021; e-mail: [email protected]
Introduction
S
everal epidemiological studies have examined the mental health status of Latino residents living in the United States. These studies include the National Institute of Mental Health (NIMH) Epidemiological Catchment Area Program,1 the Hispanic Health and Nutrition Examination Survey2 and the Hispanic Established Populations for the Epidemiologic Studies of the Elderly.3 Most of these studies were conducted among Latinos of Mexican, Cuban or Puerto-Rican descent and conducted in community-based settings. The most common measures of depressive symptoms employed in these studies included the Center for Epidemiologic Studies Depression Scale (CES-D), the Composite International Diagnostic Interview, the NIMH Diagnostic Interview Schedule and the Beck Depression Inventory.4 Irrespective of the country of origin or the measurement used, the consensus of these studies is that Latinos experience a high prevalence of depressive symptoms and that the prevalence of depressive symptoms observed among some Latinos is higher than that observed among non-Hispanic or Latino whites.4 The prevalence of depressive symptoms in these studies varied from 3–28%.5 For example, a survey of Cuban Americans living in Florida reported a lifetime prevalence of high depressive symptoms of 3%.5 Another study conducted in a cohort of community dwellers in Los Angeles County found a 27% prevalence of depresVOL. 100, NO. 7, JULY 2008 849
Race/Ethnicity and Depressive Symptoms in CAD Patients
sive symptoms among Latinos compared to a 16% prevalence in whites.6 The prevalence among Puerto Ricans living in New York City was found to be 28%.2 Several studies have also reported more adverse outcomes in patients with depressive symptoms compared to those without symptoms. Variables that have been shown to correlate the most with depressive symptoms include younger age, female gender, fewer years of education, unemployment and lower income.4 While there is ample evidence documenting the high prevalence of depressive symptoms in Latinos, there is little research on patients who have both depressive symptoms and comorbid medical conditions. One condition that deserves further investigating is coronary artery disease, the leading cause of death among Latino adults. The proportion of premature deaths attributable to coronary artery disease is estimated at 23% in Latino adults compared to 14% in white adults.7 Coronary artery disease in the Latino population is of particular concern given their high burden of cardiovascular risk factors coupled with limited socioeconomic resources and limited access to healthcare.8 Another risk factor that is understudied but that may have an equally deleterious impact on the health outcomes of Latinos with coronary artery disease is depressive symptoms. In the general population, approximately 20% of patients with coronary artery disease (CAD) are diagnosed with depressive symptoms, with 14% of patients developing new onset depressive symptoms within a year of a cardiac event.9,10 Moreover, patients with depressive symptoms are more likely to develop recurrent infarctions and more likely to undergo repeat coronary procedures than those without such symptoms.11-16 In the Hispanic Established Population for the Epidemiologic Study of the Elderly (EPESE), Latino patients who had comorbid cardiovascular disease and depressive symptoms had a mortality rate that was four times higher than patients with cardiovascular disease but who did not have depressive symptoms.17 Given the high prevalence of depressive symptoms among Latinos and the known adverse outcomes associated with such symptoms, especially in patients with coronary artery disease, depressive symptoms in Latino patients with CAD require additional research. Therefore, among Latino and white patients who recently had percutaneous transluminal coronary angioplasty (PTCA), the objectives of this study were two-fold: 1) to determine whether differences in the prevalence of depressive symptoms exist; and 2) to determine the degree to which years of education, employment status, experience of stressful life events and perceived emotional social support explained any observed differences. The hypotheses for this study are: 1) the prevalence of depressive symptoms would be higher among Latino patients compared to white patients and 2) the differences in depressive symptoms noted between Latino and white patients would be explained 850 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
by differences in education, employment status, recent stressful life events and perceived emotional social support. The rationale for these hypotheses is that there is considerable evidence to suggest that racial/ethnic categories may shape social experiences such as education, employment, social interactions and exposure to stressful events. These factors may in turn contribute the higher prevalence of depressive symptoms that has been reported among racial and ethnic minorities.18,19
Methods
This study was nested within a larger randomized controlled trial designed to improve behavior modification among patients who recently had coronary artery angioplasty. The methodologic details of the larger study have been described elsewhere.20
Patient Sample Patients were recruited from a tertiary care hospital telemetry unit and had to have angiographic evidence of coronary artery disease warranting either angioplasty and/ or stent insertion. Eligible patients who provided consent underwent baseline interviews that included an assessment of demographic characteristics (age, gender and ethnicity/race, years of education and employment status), clinical characteristics (severity of cardiac disease, medical comorbidity and quality of life) and psychosocial characteristics (depressive symptoms, perceived emotional social support and recent stressful life events).
Measures Severity of cardiac disease was measured with the Canadian Cardiovascular Society Classification system, which rates the severity of anginal symptoms from the mildest anginal symptoms (class I) to most severe symptoms (class IV).21 In addition, perceived quality of life was measured with the Medical Outcomes Study-SF36, which measures patients’ perceptions of their overall health and quality of life.22 Depressive symptoms were measured with the Center for Epidemiologic Studies Depression Scale (CESD), a 20-item measure of affective and somatic depressive symptoms with scores ranging from 0–60. A score of ≥16 has been previously validated as being indicative of having high depressive symptoms.23 Social support items explored the amount of emotional support the patient perceived such as having a friend with whom to share important and personal matters and the degree of social contact the patient had with social networks.24 Patients rated the amount of time that they had someone to talk to about personal matters from none of the time to a lot. This single-item method of assessing emotional social support has been used in prior studies.25,26 Stressful life events in the year prior to angioplasty were documented through the use of questions about family stressors, work stressors, personal stressors VOL. 100, NO. 7, JULY 2008
Race/Ethnicity and Depressive Symptoms in CAD Patients
and financial stressors. For example, participants were asked whether they had experienced the death of a family member or friend, a divorce or separation, financial troubles, troubles with children or grandchildren, a mugging, robbery, or similar events during the previous year.27 Patients were given a checklist from which they were able to select stressful life events experienced. The checklist approach allowed patients to select >1 category. Similar methodology has been used to assess stressful life events in prior studies of Latino patients.26
Data Analyses Latino categorization was based on patients’ selfidentification as either Latino or Hispanic and was contrasted with patients who identified themselves as white and non-Hispanic or non-Latino. Demographic and clinical characteristics of Latinos and whites were compared and contrasted using Chi-squared test for categorical variables and Student’s t test for continuous variables. Patients who had a score of ≥16 on the CES-D were categorized as having high depressive symptoms compared to those with a score of <16, who were categorized as having low depressive symptoms. In order to address the hypothesis that Latino patients have a higher prevalence of depressive symptoms, the proportion of depressive symptoms by race was determined. Chi square was used as a test of difference in proportions. The analytic framework for testing mediators proposed by Baron and Kenny was used to address the hypothesis that observed differences in depressive symptoms would be explained by psychosocial factors.28 Psychosocial variables that were known to be associated with both race/ethnicity and depressive symptoms were identified and tested to determine the extent to which these variables mediated the association between race/ ethnicity and depressive symptoms. According to Baron and Kenny, four conditions must be met for a variable to be considered a mediator: 1) the independent variable (race/ethnicity) must be associated with the dependent variable (high depressive symptoms); 2) the independent variable must be associated with the proposed mediators (education, employment, perceived emotional social support and recent stressful life events); 3) the proposed mediators must also be associated with the dependent variable; and 4) A test for mediation holds when the magnitude of the association between the independent and dependent variable diminishes when the proposed mediators are added to the model. The first three criteria were established with bivariate analyses. Intercorrelations between race/ethnicity and depressive symptoms, race/ethnicity and potential mediators, and potential mediators and depressive symptoms were tested with Pearson correlation. The fourth criterion was established with multivariate logistic regression models that adjusted for age, gender, severity of illness and quality of life. In each model, depressive symptoms indicated JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
by a CES-D score of ≥16 or not was the dependent variable, and race/ethnicity was the independent variable. All tests were performed using SAS version 8.29 Because many of the variables in these models were interrelated, the variance inflation factors (VIFs) for each variable was determined to test for potential of multicollinearity affecting the results. VIF measures the degree to which collinearity among the independent variables in a regression model biases associations. This was done by entering all possible independent variables into one model and calculating the VIF. A VIF of >10 indicates that multicollinearity is a problem.30 None of VIFs were >10 in these analyses.
Results
A total of 2,022 patients were identified and screened for potential eligibility through a systematic, daily review of the cardiac catheterization schedule during an 18-month period. Patients had to be admitted for PTCA or stenting. Of these patients, 1,429 were eligible, of which approximately 46% were enrolled. Patients who were eligible but were not enrolled were those who either refused or those who had a family member or physician that did not allow them to participate. There were also patients who were not enrolled because they were admitted and discharged before they could be recruited. Of the patients who were eligible but not enrolled, 13% were Latino, their mean age was 62 ± 10, and 41% were female. The remainder of the paper will focus on the 660 patients who were enrolled. Of the 660 patients who were enrolled in the parent study, 551 (59 Latino patients and 492 white patients) were included in these analyses. Table 1 shows a comparison of demographic and clinical characteristics between Latino and white patients. Latino patients were younger than white patients and had a mean age of 56 ± 10 compared to 63 ± 11 (P=0.001). Of the Latino patients, 42% were female compared to 25% of white patients, (P=0.007). Compared to white patients, fewer Latino patients were employed (P=0.001) or completed high school (P=0.001). More Latinos than whites had a history of diabetes (P= 0.003) and rated their health as fair or poor (P=0.001) on the SF-36. The mean score on the CES-D among whites was 12.6 ±11, compared to 19.2 ±15 among Latino patients (P=0.002). As shown in Table 2, 60% of Latino patients had CES-D scores of ≥16, compared to 29% of whites. Latinos were >3 times as likely to have high depressive symptoms (OR=3.5, 95% CI: 2.0–6.1). Latino patients were less likely to report that they had someone to talk to about important matters in their lives some or most of the time, compared to whites (OR=0.33, 95% CI: 15– 70). The majority of participants reported experiencing ≥1 stressful life event. Therefore, patients who reported ≥2 life events were compared with those who reported fewer events. Latinos were >3 times as likely to report VOL. 100, NO. 7, JULY 2008 851
Race/Ethnicity and Depressive Symptoms in CAD Patients
that they experienced ≥2 recent stressful life events, (OR=3.6, 95% CI: 1.9–6.9). The first three criteria of Baron and Kenny’s model were established via bivariate analyses. Pearson correlation coefficients demonstrated that the associations among race/ethnicity and depressive symptoms, race/ ethnicity and the potential mediators, and the potential mediators and depressive symptoms were all significantly correlated at the p<0.05 significance level. The final criterion was established in the multivariate models shown in Table 3, which demonstrates the impact of separately adding each of the proposed mediators on the association between race/ethnicity and depressive symptoms. As shown in Table 3, race/ethnicity was associated with depressive symptoms (OR=2.3, CI: 1.3–4.5). When the variable for recent stressful life events was added to the model, the odds ratio for the association between race/ethnicity and depressive symptoms decreased from 2.3 to 2.2 (95% CI: 1.2–4.2). When social support was added to the model as a mediator, the odds ratio for the association between race/ethnicity and depressive symptoms decreased to 2.1 (95% CI: 1.1–9.0). When education was added to the model as a mediator, the odds ratio for the association between race/ethnicity and depressive symptoms decreased to 2.0 (95% CI: 1.1–4.0). However, when employment was added to the model, the association between race/ethnicity and depressive symptoms was no longer significant—the odds ratio became 1.79 (95% CI: 0.75–4.2). All models included age, gender, severity of illness and quality of life.
Discussion
This study applied Baron and Kenny’s framework to
identify potential mediators in the association between race/ethnicity and depressive symptoms. Of the Latino patients in this study, 60% had high depressive symptoms, compared to 29% of whites, which established the criteria in Baron and Kenny’s model that the independent variable must be associated with the dependent variable. Bivariate associations between race/ethnicity and the potential mediators and between the potential mediators and depressive symptoms were also established. The final criterion for mediation was confirmed in logistic regression models that showed that the association between racial/ethnicity and depressive symptoms was partially explained by fewer years of education, being unemployed, lacking emotional social support and experiencing ≥2 recent stressful life events. Of these variables, unemployment had the greatest impact on the association between race/ethnicity and depressive symptoms. This is relevant because employment status may influence access to care and the ability to manage one’s health. This is especially important since Latinos in this study were younger, more likely to be female and unmarried. In this context, employment status would have an important impact on the patients’ ability to care for themselves and perhaps their family. The fact that there were more female participants among Latinos compared to whites is worth further mention. It has been documented that there is a greater prevalence of depressive symptoms among women than men in both the general population and in Latino populations.31,32 It is also known that among patients with coronary artery disease, women are twice as likely to have depressive symptoms.33 Although gender did not affect the association between depressive symptoms and race/ ethnicity, depressive symptoms in Latina women are of
Table 1. Clinical characteristics of Latino patients compared to white patients
Demographic Characteristics Age (years) + SD Female Married Employed High-school graduate Clinical History Diabetes Self-rated health as fair or poor Hypertension history Unstable angina Previous infarction Previous heart failure history Severity of illness (CCS) class IV*
Latino (n=59)
White (n=601)
P
56 ± 10 42% 56% 44% 58%
63 ± 11 25% 63% 56% 88%
0.001 0.007 0.001 0.001 0.001
43% 64% 58% 66% 27% 15% 57%
24% 29% 54% 50% 24% 10% 47%
0.003 0.001 NS* NS NS NS NS
* Canadian Cardiovascular Society Classification; ** NS: Nonsignificant, indicates that p>00.05
852 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
VOL. 100, NO. 7, JULY 2008
Race/Ethnicity and Depressive Symptoms in CAD Patients
particular concern. According to traditional sex roles that define many Latino households, it is common for Latina women to have to tend for the household needs of the family such as caring for the children and their partner.34 Latina females in this study may have had to maintain their roles in the family while managing both coronary artery disease and depressive symptoms. Future studies focusing on depressive symptoms among Latina women may help to further explore this assumption. These findings must also be viewed in light of what is known about the socioeconomic status of Latinos living in the United States. According to the Department of Health and Human Services 2000 Census report, some subgroups within the Latino population may have fewer socioeconomic resources in terms of years of education, income and employment that are available to them.35 Education and employment are economic resources that are necessary to cope effectively with day-to-day stressors. Thus, patients who lack or are bereft of such resources may be at greater risk for depressive symptoms.36 The finding that Latino patients reported lower perceived social support is surprising, given that social support—in particular, family support—has consistently been shown to be a protective factor within Latino families.26 The cultural value of familismo has been used to describe the strong attachments to extended and nuclear families.37,38 However, the role of social support in patients with comorbid illnesses is unknown. It is possible that the larger family networks may not be able to adequately address the needs of family members who have comorbid medical conditions, especially if socioeconomic resources are limited. While there may be other limitations in this study, we
will focus on four limitations that were felt to be most germane to interpreting and extrapolating our results. These limitations occur in terms of the number of Latinos recruited, the selection of study participants, the measurement of social support and stressful life events and the method used to test for mediation. First, there were too few Latinos to allow for within Latino comparisons of those patients who had high depressive symptoms and those that did not. Second, common to studies in clinically predefined populations, the selection of study participant limits the ability to generalize the findings to other Latinos who do not have coronary artery disease. The requirement that patients spoke English also limits the generalizability of these findings and the number of Latino participants in the study. Future studies can address these limitations and build upon our findings by recruiting a larger number of Latinos patients with other comorbid conditions, who have diverse language fluency, and by also capturing information on their country of origin and level of acculturation. This would allow for a more in-depth description of the tremendous heterogeneity that exists within the Latino population. Third, the use of a single item to measure emotional social support is another limitation of the study. Compared to multi-item measures, a single item precludes the ability to evaluate other aspects of social support such as informational and tangible support. However, of all of the categories of social support, emotional social support has consistently been shown to predict health outcomes. Previous studies have used a similar item to measure emotional social support and found it be predictive of mortality after an infarction, fatal and nonfatal events in the setting of heart failure and intention to participate
Table 2. Psychosocial characteristics by race/ethnicity
Psychosocial Characteristics Scored ≥16 on the CES-D Had someone to talk to some or most of the time Number of recent life events
Latino (n=59) 60% 81% 28%
White (n=601) 29% 92% 9%
Odds Ratio for Latino (95% CI) 3.5 (2.0,6.1) 0.33 (0.15, 0.70) 3.6 (1.9,6.9)
Table 3. Multivariate models demonstrating the impact of adding mediators on the association between race/ethnicity on depressive symptoms
Model Race/ethnicity Stressful life event Emotional social support Education Employment
Odds Ratio
95% CI
2.3 2.2 2.1 2.0 1.8
1.3-4.5 1.2-4.2 1.1-9.0 1.1-4.0 0.75-4.2
* All models adjusted for age, gender, severity of illness.
JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
VOL. 100, NO. 7, JULY 2008 853
Race/Ethnicity and Depressive Symptoms in CAD Patients
in healthy behaviors.25,39,40 The checklist approach to measuring life events has been commonly used in prior studies, but it does not allow for a more detailed evaluation of the specific categories of events. In this study, some respondents experienced >1 event, and this approach allowed them to select >1 stressful event. However, because the duration of the event was not probed, the impact that duration of life events may have had on depressive symptoms could not be determined. This is an important consideration for future studies since the presence of chronic stress may represent unresolved issues that continually cause distress. Finally, a series of logistic regression models was used to test for mediation using Baron and Kenny’s approach as a framework. This is a valid method; however, it limited the ability to identify direct and indirect pathways leading to symptoms of depressive symptoms. Despite these limitations, this study contributes to an understanding of ways in which psychosocial variables may influence mental health in Latino patients with coronary artery disease. Examining these variables in the context of coronary artery disease is important in light of three contemporary trends that have been observed. These are: the increasing number of Latinos in the United States; the increasing prevalence of depressive symptoms in the general population; and the increasing prevalence in cardiovascular risk factors, including obesity, physical inactivity and type-2 diabetes.41-44 These trends make it highly probable that an increase in both the prevalence of depressive symptoms and coronary artery disease will be observed in the Latino population. This probability is further complicated by the lower use of mental health services among Latinos.45,46 Therefore, Public health initiatives to screen for depression and to provide referrals are important to addressing mental health in Latinos, in particular, those with coronary artery disease. In conclusion, our findings do not suggest that Latinos have an inherent negative disposition but, rather, illustrate the importance of understanding the factors that may influence the presence of depressive symptoms. Race/ethnicity may be a proxy measure for a set of interrelated social experiences such as access to education, employment, social interactions and exposure to stressful events. Reducing racial/ethnic disparities in mental health may require public health initiatives that address these psychosocial factors.
References
1. Golding JM, Lipton RI. Depressed mood and major depressive disorder in two ethnic groups. J Psychiatr Res. 1990;24(1):65-82. 2. Potter LB, Rogler LH, Moscicki EK. Depression among Puerto Ricans in New York City: the Hispanic Health and Nutrition Examination Survey. Soc Psychiatry Psychiatr Epidemiol. 1995;30(4):185-193. 3. Black SA, Markides KS, Miller TQ. Correlates of depressive symptomatology among older community-dwelling Mexican Americans: the Hispanic EPESE. J Gerontol B Psychol Sci Soc Sci. 1998;53(4):S198-208. 4. Saez-Santiago E, Bernal G. Depression in ethnic minorities: Latinos and
854 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
Latinas, African Americans, Asian Americans, and Native Americans. Handbook of racial & ethnic minority psychology. Vol 4. Thousand Oaks, CA: Sage; 2003:401-428. 5. Narrow WE, Rae DS, Moscicki EK, et al. Depression among Cuban Americans. The Hispanic Health and Nutrition Examination Survey. Soc Psychiatry Psychiatr Epidemiol. 1990;25(5):260-268. 6. Frerichs RR, Aneshensel CS, Clark VA. Prevalence of depression in Los Angeles County. Am J Epidemiol. 1981;113(6):691-699. 7. Thom T, Haase N, Rosamond W, et al. Heart disease and stroke statistics—2006 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2006;113(6): e85-e151. 8. Sundquist J, Winkleby MA, Pudaric S. Cardiovascular disease risk factors among older black, Mexican-American, and white women and men: an analysis of NHANES III, 1988–1994. Third National Health and Nutrition Examination Survey. J Am Geriatr Soc. 2001;49(2):109-116. 9. Mallik S, Krumholz HM, Lin ZQ, et al. Patients with depressive symptoms have lower health status benefits after coronary artery bypass surgery. Circulation. 2005;111(3):271-277. 10. Lauzon C, Beck CA, Huynh T, et al. Depression and prognosis following hospital admission because of acute myocardial infarction. CMAJ. 2003;168(5):547-552. 11. Borowicz L, Royall R, Grega M, et al. Depression and cardiac morbidity 5 years after coronary artery bypass surgery. Psychosomatics. 2002;43(6):464471. 12. Burker EJ, Blumenthal JA, Feldman M, et al. Depression in male and female patients undergoing cardiac surgery. Br J Clin Psychol. 1995;34(Pt 1):119-128. 13. Carney RM, Freedland KE, Veith RC, et al. Major depression, heart rate, and plasma norepinephrine in patients with coronary heart disease. Biol Psychiatry. 1999;45(4):458-463. 14. Carney RM, Freedland KE, Eisen SA, et al. Major depression and medication adherence in elderly patients with coronary artery disease. Health Psychol. 1995;14(1):88-90. 15. Carney RM, Rich MW, Freedland KE, et al. Major depressive disorder predicts cardiac events in patients with coronary artery disease. Psychosom Med. 1988;50(6):627-633. 16. Musselman DL, Evans DL, Nemeroff CB. The relationship of depression to cardiovascular disease: epidemiology, biology, and treatment. Arch Gen Psychiatry. 1998;55(7):580-592. 17. Black SA, Markides KS. Depressive symptoms and mortality in older Mexican Americans. Ann Epidemiol. 1999;9(1):45-52. 18. Turner RJ, Lloyd DA. The stress process and the social distribution of depression. J Health Soc Behav. 1999;40(4):374-404. 19. Schulz AJ, Israel BA, Zenk SN, et al. Psychosocial stress and social support as mediators of relationships between income, length of residence and depressive symptoms among African American women on Detroit’s eastside. Soc Sci Med. 2006;62(2):510-522. 20. Charlson ME, Allegrante JP, McKinley PS, et al. Improving health behaviors and outcomes after angioplasty: using economic theory to inform intervention. Health Educ Res. 2002;17(5):606-618. 21. Campeau L. The Canadian Cardiovascular Society grading of angina pectoris revisited 30 years later. Can J Cardiol. 2002;18(4):371-379. 22. Ware JE, Kosinski M, and Keller SD. SF-36 Physical and Mental Health Summary Scales: A User’s Manual. Boston: The Health Institute, New England Medical Center; 1994. 23. Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385-401. 24. Gorkin L, Schron EB, Brooks MM, et al. Psychosocial predictors of mortality in the Cardiac Arrhythmia Suppression Trial-1 (CAST-1). Am J Cardiol. 1993;71(4):263-267. 25. Berkman LF, Leo-Summers L, Horwitz RI. Emotional support and survival after myocardial infarction. A prospective, population-based study of the elderly. Ann Intern Med. 1992;117(12):1003-1009. 26. Chiriboga DA, Black SA, Aranda M, et al. Stress and depressive symptoms among Mexican American elders. J Gerontol B Psychol Sci Soc Sci. 2002;57(6):P559-568. 27. Ahern DK, Gorkin L, Anderson JL, et al. Biobehavioral variables and mor-
VOL. 100, NO. 7, JULY 2008
Race/Ethnicity and Depressive Symptoms in CAD Patients tality or cardiac arrest in the Cardiac Arrhythmia Pilot Study (CAPS). Am J Cardiol. 1990;66(1):59-62. 28. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality & Social Psychology. 1986;51(6):1173-1182. 29. SAS Institute Inc., SAS OnlineDoc®, Version 8, Cary, NC: SAS Institute Inc., 2000. 30. Tucker JL. The moderators of patient satisfaction. J Manag Med. 2002;16:48-66. 31. Salgado de Snyder VN, Cervantes RC, Padilla AM. Gender and ethnic differences in psychosocial stress and generalized distress among Hispanics. Sex Roles. 1990;22(7-8):441-453. 32. Koss-Chioino JD. Depression among Puerto Rican Women: Culture, Etiology and Diagnosis. Hispanic Journal of Behavioral Sciences. 1999;21(3):330-350. 33. Frasure-Smith N, Lesperance F, Juneau M, et al. Gender, depression, and one-year prognosis after myocardial infarction. Psychosom Med. 1999;61(1):26-37. 34. Miranda AO, Bilot JM, Peluso PR, et al. Latino Families: the Relevance of the Connection Among Acculturation, Family Dynamics, and Health for Family Counseling Research and Practice. The Family J. 2006;14(3): 268273. 35. Ramirez R. We the People: Hispanics in the United States. www.census. gov/prod/2004pubs/censr. Accessed 05/06/08. 36. Plant EA, Sachs-Ericsson N. Racial and ethnic differences in depression: the roles of social support and meeting basic needs. J Consult Clin Psychol. 2004;72(1):41-52. 37. La Roche MJ. The Association of Social Relations and Depression Levels among Dominicans in the United States. Hispanic J Behav Sci. 1999;21(4):420-430.
c
a
r
e
e
r
o
p
p
o
r
JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION
t
38. Losada A, Robinson Shurgot G, Knight BG, et al. Cross-cultural study comparing the association of familism with burden and depressive symptoms in two samples of Hispanic dementia caregivers. Aging Ment Health. 2006;10(1):69-76. 39. Rhodes RE, Jones LW, Courneya KS. Extending the theory of planned behavior in the exercise domain: a comparison of social support and subjective norm. Res Q Exerc Sport. 2002;73(2):193-199. 40. Krumholz HM, Butler J, Miller J, et al. Prognostic importance of emotional support for elderly patients hospitalized with heart failure. Circulation. 1998;97(10):958-964. 41. Larkin GL, Claassen CA, Emond JA, et al. Trends in U.S. emergency department visits for mental health conditions, 1992 to 2001. Psychiatr Serv. 2005;56(6):671-677. 42. Klerman GL, Weissman MM. Increasing rates of depression. JAMA. 1989;261(15):2229-2235. 43. Arnett DK, McGovern PG, Jacobs DR, et al. Fifteen-year trends in cardiovascular risk factors (1980-1982 through 1995-1997): the Minnesota Heart Survey. Am J Epidemiol. 2002;156(10):929-935. 44. Lorenzo C, Okoloise M, Williams K, et al. The metabolic syndrome as predictor of type 2 diabetes: the San Antonio heart study. Diabetes Care. 2003;26(11):3153-3159. 45. Alegria M, Canino G, Rios R, et al. Inequalities in use of specialty mental health services among Latinos, African Americans, and non-Latino whites. Psychiatr Serv. 2002;53(12):1547-1555. 46. Miranda J, Cooper LA. Disparities in care for depression among primary care patients. J Gen Intern Med. 2004;19(2):120-126. n
u
n
i
t
i
e
s
VOL. 100, NO. 7, JULY 2008 855